Transforming medicine through AI-enabled healthcare pathways

Advances in machine learning are set to transform medicine, yet considerable challenges exist.  These include hard technical ones such as the complexity and quality of healthcare data, the need to consider multiple interactions between the diverse events over a patient’s life and the difficulty of estimating counterfactual outcomes.  These challenges are exacerbated by practical ones – the requirement for cross-disciplinary collaboration, the delivery and transferability of any solutions between health systems at scale both nationally and internationally, the deployment of a service that provides tailored patient-level intelligence in near-real-time and most fundamentally, the absence of a way to conceptualise the complexity of healthcare to support interdisciplinary collaboration. 

The speakers in this talk have joined forces to work on these challenges.  We have combined a world-leading team in machine learning with the world’s largest, near-real time, high-quality cancer data collection service.  Together we have built a test and development environment to create new methods for the near-real-time analysis of patient-level cancer data, that once tested, can then be rapidly integrated into the same national data collection service to benefit all patients. 

In this talk we will explain the background to our work and demonstrate how we have addressed some of these challenges.  By using cancer as an exemplar, we aim to create a system-theoretical approach to healthcare that will help facilitate a deeper understanding of medicine and foster rigorous interdisciplinary working.

Agenda